Add prompting frameworks guide and expand prompt documentation#1073
Add prompting frameworks guide and expand prompt documentation#1073Krosebrook wants to merge 404 commits intof:mainfrom
Conversation
Co-authored-by: thanos0000@gmail.com <thanos0000@gmail.com@users.noreply.github.com>
Co-authored-by: gem00cem@gmail.com <gem00cem@gmail.com@users.noreply.github.com>
Includes comprehensive documentation for RTF, RISEN, CO-STAR, CRISPE, APE, RACE, ROSES, RASCEF, Chain of Thought, TAG, BAB, CARE, SCAMPER, TRACE, and ERA frameworks. Each framework has a template and 3 practical examples covering various use cases.
|
No actionable comments were generated in the recent review. 🎉 ℹ️ Recent review info⚙️ Run configurationConfiguration used: Organization UI Review profile: CHILL Plan: Pro Run ID: 📒 Files selected for processing (1)
✅ Files skipped from review due to trivial changes (1)
📝 WalkthroughWalkthroughAdds a new documentation guide, Changes
Estimated code review effort🎯 2 (Simple) | ⏱️ ~10 minutes Poem
🚥 Pre-merge checks | ✅ 3✅ Passed checks (3 passed)
✏️ Tip: You can configure your own custom pre-merge checks in the settings. ✨ Finishing Touches🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
There was a problem hiding this comment.
Pull request overview
This PR adds a new PROMPTING-FRAMEWORKS.md file documenting 15 prompt engineering frameworks with templates and examples, and adds a link to it from the README.
Changes:
- New
PROMPTING-FRAMEWORKS.mdwith 15 prompting frameworks (RTF, RISEN, CO-STAR, CRISPE, APE, RACE, ROSES, RASCEF, Chain of Thought, TAG, BAB, CARE, SCAMPER, TRACE, ERA), each with a template and 3 practical examples, plus a selection guide and tips - Updated
README.mdwith a link to the new frameworks guide
Reviewed changes
Copilot reviewed 2 out of 4 changed files in this pull request and generated 3 comments.
| File | Description |
|---|---|
| README.md | Added a link to the new PROMPTING-FRAMEWORKS.md guide |
| PROMPTING-FRAMEWORKS.md | New comprehensive guide covering 15 prompting frameworks with templates, examples, and a selection guide |
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
You can also share your feedback on Copilot code review. Take the survey.
PROMPTING-FRAMEWORKS.md
Outdated
| ```json | ||
| POST /api/v1/users | ||
| { | ||
| "email": "user@example.com", | ||
| "password": "securePassword123" | ||
| } | ||
| ``` |
PROMPTING-FRAMEWORKS.md
Outdated
| ```yaml | ||
| name: CI/CD Pipeline | ||
| on: | ||
| push: | ||
| branches: [main, develop] | ||
| ``` |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
🧹 Nitpick comments (1)
PROMPTING-FRAMEWORKS.md (1)
35-1212: Consider adding language specifiers to fenced code blocks.The framework templates in code blocks currently lack language specifiers, which triggers markdown linting warnings. While these blocks contain prompt templates rather than programming code, adding a language identifier would improve markdown compliance and accessibility.
Consider adding
textormarkdownas the language identifier:```text **Role:** [Who should the AI act as] **Task:** [What you want the AI to do] **Format:** [How the output should be structured] ```This applies to template blocks throughout the document (69 instances flagged by markdownlint).
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@PROMPTING-FRAMEWORKS.md` around lines 35 - 1212, The fenced template code blocks (those starting with ``` and containing lines like "**Role:**", "**Task:**", "**Format:**" and the RISEN/CO-STAR/etc. examples) lack language specifiers and trigger markdownlint; update each template code block (all instances of the triple-backtick blocks that contain the prompt templates) to use a language identifier such as ```text or ```markdown so linting passes and accessibility improves; ensure you update every occurrence (~69 instances flagged) consistently across the document while preserving the exact block content.
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Nitpick comments:
In `@PROMPTING-FRAMEWORKS.md`:
- Around line 35-1212: The fenced template code blocks (those starting with ```
and containing lines like "**Role:**", "**Task:**", "**Format:**" and the
RISEN/CO-STAR/etc. examples) lack language specifiers and trigger markdownlint;
update each template code block (all instances of the triple-backtick blocks
that contain the prompt templates) to use a language identifier such as ```text
or ```markdown so linting passes and accessibility improves; ensure you update
every occurrence (~69 instances flagged) consistently across the document while
preserving the exact block content.
ℹ️ Review info
⚙️ Run configuration
Configuration used: Organization UI
Review profile: CHILL
Plan: Pro
Run ID: b5a49115-00da-4aa8-8809-fb23ff251095
⛔ Files ignored due to path filters (1)
prompts.csvis excluded by!**/*.csv
📒 Files selected for processing (3)
PROMPTING-FRAMEWORKS.mdPROMPTS.mdREADME.md
|
@copilot open a new pull request to apply changes based on the comments in this thread |
There was a problem hiding this comment.
Actionable comments posted: 1
🧹 Nitpick comments (2)
PROMPTING-FRAMEWORKS.md (2)
85-86: Source or soften specific factual claims about framework origins/awards.Statements like “Created by …” and “won … competition” should include a citation (or be reworded as “often attributed to”) to keep the guide durable and verifiable.
Also applies to: 176-177
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@PROMPTING-FRAMEWORKS.md` around lines 85 - 86, Replace definitive origin/award claims in PROMPTING-FRAMEWORKS.md—e.g., the sentence fragment "Created by Kyle Balmer." and any "won … competition" claims (also present around lines 176-177)—with either a concise citation to a reliable source or a softened phrasing such as "often attributed to Kyle Balmer" or "reported to have won [competition]" and add an inline reference note or footnote; ensure the exact phrases "Created by Kyle Balmer." and the award wording are updated so every factual claim is either sourced or clearly hedged.
35-1211: Add language identifiers to fenced code blocks to resolve MD040 warnings.Most fenced blocks use plain triple backticks without a language, which triggers markdownlint repeatedly and hurts editor rendering consistency.
Suggested pattern
-``` +```text **Role:** [Who should the AI act as] **Task:** [What you want the AI to do] **Format:** [How the output should be structured] -``` +```Apply this consistently to all prompt-template/example fences (or use
markdownwhere you want nested markdown highlighting).🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@PROMPTING-FRAMEWORKS.md` around lines 35 - 1211, The markdown contains many fenced code blocks (e.g., the block beginning with "**Role:** [Who should the AI act as]" and all example/templates like RISEN, CO-STAR, etc.) that lack language identifiers and trigger markdownlint MD040; update every triple-backtick fence to include an explicit language token (use "text" for plain prompt templates or "markdown" where nested markdown highlighting is desired) so all examples and template fences are consistent, then re-run linting to confirm MD040 is resolved across the document.
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Inline comments:
In `@PROMPTING-FRAMEWORKS.md`:
- Around line 418-450: Add an explicit safety disclaimer to the financial
planning and medical examples by inserting a short “Informational only — consult
a licensed professional” note matching the legal example’s pattern;
specifically, append a one-sentence disclaimer after the Expectations block in
the CFP example (the "**Role:** You are a Certified Financial Planner (CFP)..."
/ savings framework) and after the Medical Research Synthesis example (the
"**Role:** You are a clinical researcher..." block), and apply the same change
to the duplicate block referenced at lines 1145-1163 so both examples include
the explicit safety disclaimer.
---
Nitpick comments:
In `@PROMPTING-FRAMEWORKS.md`:
- Around line 85-86: Replace definitive origin/award claims in
PROMPTING-FRAMEWORKS.md—e.g., the sentence fragment "Created by Kyle Balmer."
and any "won … competition" claims (also present around lines 176-177)—with
either a concise citation to a reliable source or a softened phrasing such as
"often attributed to Kyle Balmer" or "reported to have won [competition]" and
add an inline reference note or footnote; ensure the exact phrases "Created by
Kyle Balmer." and the award wording are updated so every factual claim is either
sourced or clearly hedged.
- Around line 35-1211: The markdown contains many fenced code blocks (e.g., the
block beginning with "**Role:** [Who should the AI act as]" and all
example/templates like RISEN, CO-STAR, etc.) that lack language identifiers and
trigger markdownlint MD040; update every triple-backtick fence to include an
explicit language token (use "text" for plain prompt templates or "markdown"
where nested markdown highlighting is desired) so all examples and template
fences are consistent, then re-run linting to confirm MD040 is resolved across
the document.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Organization UI
Review profile: CHILL
Plan: Pro
Run ID: 7bd3c82b-2bd7-4cf9-9038-ec6e35029643
📒 Files selected for processing (1)
PROMPTING-FRAMEWORKS.md
- Add disclaimers to financial planning and medical examples - Soften 'Created by' to 'Often attributed to' for RISEN framework - Change 'won' to 'reportedly won' for CO-STAR competition claim Addresses CodeRabbit review comments.
There was a problem hiding this comment.
🧹 Nitpick comments (1)
PROMPTING-FRAMEWORKS.md (1)
35-35: Add language identifiers to fenced code blocks to satisfy markdown lint.All fenced blocks are missing a language tag (
MD040), which can fail docs CI. Usetext(ormarkdownwhere appropriate) consistently.Suggested pattern (apply globally)
-``` +```text **Role:** [Who should the AI act as] **Task:** [What you want the AI to do] **Format:** [How the output should be structured]</details> Also applies to: 43-43, 53-53, 67-67, 89-89, 99-99, 118-118, 147-147, 180-180, 191-191, 211-211, 230-230, 258-258, 268-268, 288-288, 306-306, 332-332, 340-340, 354-354, 369-369, 391-391, 400-400, 417-417, 436-436, 464-464, 474-474, 495-495, 516-516, 545-545, 556-556, 587-587, 622-622, 660-660, 667-667, 684-684, 700-700, 737-737, 763-763, 771-771, 786-786, 806-806, 829-829, 837-837, 851-851, 865-865, 887-887, 896-896, 908-908, 921-921, 947-947, 961-961, 1004-1004, 1047-1047, 1098-1098, 1108-1108, 1127-1127, 1146-1146, 1179-1179, 1187-1187, 1197-1197, 1207-1207 <details> <summary>🤖 Prompt for AI Agents</summary>Verify each finding against the current code and only fix it if needed.
In
@PROMPTING-FRAMEWORKS.mdat line 35, The markdown file
PROMPTING-FRAMEWORKS.md has many fenced code blocks missing language identifiers
(MD040); update each triple-backtick block to include a language tag — use
text for plain examples ormarkdown where the block contains markdown —
following the suggested pattern (e.g., the Role/Task/Format examples); apply
this change consistently to all indicated fenced blocks so CI lint no longer
flags MD040.</details> </blockquote></details> </blockquote></details> <details> <summary>🤖 Prompt for all review comments with AI agents</summary>Verify each finding against the current code and only fix it if needed.
Nitpick comments:
In@PROMPTING-FRAMEWORKS.md:
- Line 35: The markdown file PROMPTING-FRAMEWORKS.md has many fenced code blocks
missing language identifiers (MD040); update each triple-backtick block to
include a language tag — usetext for plain examples ormarkdown where the
block contains markdown — following the suggested pattern (e.g., the
Role/Task/Format examples); apply this change consistently to all indicated
fenced blocks so CI lint no longer flags MD040.</details> --- <details> <summary>ℹ️ Review info</summary> <details> <summary>⚙️ Run configuration</summary> **Configuration used**: Organization UI **Review profile**: CHILL **Plan**: Pro **Run ID**: `a7b07518-40e8-40dd-b550-8ae3b12b3c2c` </details> <details> <summary>📥 Commits</summary> Reviewing files that changed from the base of the PR and between e25c41b07b8eaf690a30d03c5cb7af419ee1b7fc and bd1574d5cce2f8abf9af538e1102e27f8f97595d. </details> <details> <summary>📒 Files selected for processing (1)</summary> * `PROMPTING-FRAMEWORKS.md` </details> </details> <!-- This is an auto-generated comment by CodeRabbit for review status -->
Addresses MD040 markdown lint warning for missing language specifiers.
|
@copilot apply changes based on the comments in this thread |
|
Well, we don't update prompts.csv and prompts.md directly. You must use prompts.chat to create prompts. These are only mirrors. |
Description
This PR adds comprehensive documentation for prompt engineering frameworks and expands the existing prompt collection:
PROMPTING-FRAMEWORKS.md- A detailed guide covering 15 popular prompting frameworks (RTF, RISEN, CO-STAR, CRISPE, APE, RACE, ROSES, RASCEF, Chain of Thought, TAG, SPADE, SCAMPER, CLEAR, AIDA, and PREP) with templates and practical examples for eachPROMPTS.md- Significantly expanded with additional prompt content and examplesprompts.csv- Extended prompt collection with new entriesREADME.md- Added references to the new frameworks guideType of Change
Please don't edit
prompts.csvdirectly!Instead, visit prompts.chat and:
This ensures proper attribution, formatting, and keeps the repository in sync. You'll also appear on the Contributors page!
Additional Notes
The new
PROMPTING-FRAMEWORKS.mdfile provides a valuable resource for users looking to improve their prompt engineering skills with well-documented, battle-tested frameworks and real-world examples.https://claude.ai/code/session_015H9AvSBVn9EtEPV9Ccz72u
Summary by CodeRabbit